home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!saimiri.primate.wisc.edu!sdd.hp.com!think.com!barmar
- From: barmar@think.com (Barry Margolin)
- Newsgroups: comp.sys.sun.admin
- Subject: Re: libc.so.* and libc.sa.*: kernel arch dependent?
- Date: 17 Dec 1992 03:59:31 GMT
- Organization: Thinking Machines Corporation, Cambridge MA, USA
- Lines: 24
- Message-ID: <1gotv3INNeqr@early-bird.think.com>
- References: <1992Dec16.224530.21853@rigel.econ.uga.edu>
- NNTP-Posting-Host: gandalf.think.com
-
- In article <1992Dec16.224530.21853@rigel.econ.uga.edu> glenn@rigel.econ.uga.edu (Glenn F. Leavell) writes:
- >While rebuilding /usr/lib/libc.so.* and /usr/lib/libc.sa.*, an interesting
- >question has come up. These libraries are, of course, application architecture
- >dependent, but are they kernel architecture dependent?
-
- No. If they were, they'd be in /usr/kvm rather than /usr/lib (there would
- be links in /usr/lib, of course).
-
- Consider: the shared libraries are compiled from the same sources and
- headers as the static libraries, and programs linked statically with the
- latter are portable between different kernel architectures.
-
- In general, programs are only kernel-dependent if they access /dev/mem or
- /dev/kmem, and such programs have to be setuid. Since libraries can't be
- setuid, they shouldn't try to access them. (Strictly speaking, this rule
- could be violated for those system functions that are documented as being
- restricted to the superuser, since a user-mode implementation might work by
- accessing /dev/mem -- however, this isn't how any Unix system calls are
- implemented.)
- --
- Barry Margolin
- System Manager, Thinking Machines Corp.
-
- barmar@think.com {uunet,harvard}!think!barmar
-